Methods
This study was a Phase 2b, multicenter, international double-blind, randomized, placebo-controlled, parallel group trial, conducted in compliance with the ethical principles of the Declaration of Helsinki and in compliance with all International Conference on Harmonization Good Clinical Practice Guidelines. Ethical approval was obtained from the Ethics Committees of each participating center, and written informed consent was obtained from all patients or their legal representative.
Adult patients (>18 yr) admitted to one of the 233 study centers in 17 countries (Appendix 1, Supplemental Digital Content 1, http://links.lww.com/CCM/A674) with sepsis, defined according to the American College of Chest Physicians/Society of Critical Care Medicine consensus definition, and suspected DIC, identified locally at each center using a score based on platelet count and prothrombin time (PT)-international normalized ratio (INR) derived from the International Society on Thrombosis and Hemostasis (ISTH) DIC score, were screened for inclusion. There were four protocol amendments during the study period (v 1.0–1.4); changes from 1.1 to 1.2 and from 1.3 to 1.4 affected the DIC algorithm, the first to expand enrollment and the second to increase the yield of subjects with DIC (Table 1). The numbers of patients enrolled in each amendment are shown in Figure 1.
(Enlarge Image)
Figure 1.
Flow of patients through the study and numbers enrolled in each of the protocol amendment groups.
Patients were excluded from the study if any of the following criteria were present (full details are provided in Appendix 2, Supplemental Digital Content 1, http://links.lww.com/CCM/A674): unable to obtain informed consent; presence of any disorder other than sepsis that could alter coagulation; recent history of significant bleeding or increased risk of bleeding (e.g., surgery within 12 hr of screening); presence of a disorder requiring anticoagulation; use of drotrecogin alfa (activated) within the 24 hours prior to enrollment or intended use; use of anticoagulants, antiplatelet agents, antithrombotics, and thrombolytics within the 24 hours prior to study dosing, or intended use, with the exception of heparin locks/flushes or deep-vein thrombosis prophylaxis; platelet count < 20×10/L; life expectancy < 90 days or current use of any chemotherapy agent.
After enrollment, patients were randomly assigned in a 1:1 ratio, using a dynamic interactive web response system/interactive voice response system system, with the country (n = 17) and the modified disseminated intravascular coagulation (mDIC) algorithm (=2 or >2) as stratification factors, to receive either IV ART-123 (0.06 mg/kg/d) for 6 days or placebo, in addition to standard of care as determined by the treating clinician. ART-123 and placebo were supplied in 1 mL glass ampoules, and patients and staff were blinded to their contents. The dose of ART-123 administered was 0.06 mg/kg (0.01 mL/kg) up to a maximum dose of 6 mg (1 mL) for patients who weighed more than 100 kg. The study drug (ART-123 or placebo) was administered by either IV bolus injection or IV infusion (diluted in 50 mL 0.9% saline) over 15 minutes, via a dedicated IV catheter.
Patients were followed for 28 days after study inclusion. Recorded baseline characteristics included demographic information and information on preexisting conditions and treatments, organ function, and hematologic and other laboratory tests. Central laboratory evaluations for the full ISTH algorithm DIC score (Table 2) were performed at randomization (predose), days 1, 3 (predose and postdose), 7, and 14; overt DIC was diagnosed when the ISTH score was ≥ 5. Blood samples were taken for serum chemistry and hematology determinations on days 3, 7, 14, and 28. The presence of ART-123 antibodies was assessed using an enzyme-linked immunosorbent assay method at baseline and on days 7, 14, and 28. All samples determined to be positive for antibody to ART-123 on day 28 were tested for the presence of neutralizing antibodies. Patients with antibodies present on day 28 were followed until antibody was no longer detected.
Plasma concentrations of ART-123 (ng/mL) were measured using enzyme-linked immunosorbent assay on day 1 (12 hr postdose), day 3 (predose and 12 ± 3 hr postdose), day 7, and day 14. Blood samples were also obtained at these time points to measure coagulation (TATc, prothrombin fragment F1.2, protein C, protein C inhibitor, plasminogen activator inhibitor-1], antithrombin III) and inflammatory (C5a, interleukin 6, interleukin 10, myeloperoxidase, procalcitonin, high-sensitivity C-reactive protein) markers. Presence of organ dysfunction for four organ systems (renal: creatinine >2 mg/dL; hepatic: bilirubin > 2.0 mg/dL; respiratory: use of mechanical ventilation; cardiovascular: use of a vasopressor) was assessed daily until patients were discharged from hospital. Creatinine clearance was estimated using the Cockcroft-Gault equation. Pharmacokinetic variables, including volume of distribution (V) and drug clearance (CL), were estimated using a one-compartment model.
Information about serious adverse events (SAEs) was collected prospectively and recorded. An SAE was considered any undesirable sign, symptom, or medical condition that was fatal or life threatening, required prolonged inpatient hospitalization or rehospitalization, resulted in persistent or significant disability/incapacity, was medically significant as determined by a qualified health professional, or was a major bleeding event.
Statistical Analyses
The primary endpoint for this study, based on its clinical relevance, was reduction in mortality. Secondary endpoints included reversal of overt DIC and indicators of reduction in disease severity, including shock-free and alive days, ventilator-free and alive days, and dialysis-free and alive days. Event-free and alive days were calculated based on the number of days a patient was known to be alive and free of the event through day 28.
The study was powered for an endpoint of 28-day all-cause mortality. A sample size of 750 patients was sufficient to provide 90% power at a 5% two-sided alpha level based on an assumption of a placebo mortality rate of 32%, ART-123 mortality rate of 21%, and missing vital status rate of 2.5% (patients with unknown vital status at day 28 were classified as dead for the primary analysis). A prespecified blinded interim analysis conducted after 100 patients had been enrolled showed that the actual prevalence of DIC was insufficient to meet these assumptions. Hence, with the consent of the data-monitoring committee, the study design was changed (protocol amendment 3) to that of a screening trial and decision rules introduced that defined both a "definitive" level of evidence (one-sided p values < 2.5%), indicating definite efficacy, and a "suggestive" level of evidence (one-sided p values < 15%), indicating possible efficacy and warranting further investigation.